mariadb/storage/myisammrg/myrg_close.c
unknown e223c32077 Bug#30273 - merge tables: Can't lock file (errno: 155)
The patch for Bug 26379 (Combination of FLUSH TABLE and
REPAIR TABLE corrupts a MERGE table) fixed this bug too.
However it revealed a new bug that crashed the server.

Flushing a merge table at the moment when it is between open
and attach of children crashed the server.

The flushing thread wants to abort locks on the flushed table.
It calls ha_myisammrg::lock_count() and ha_myisammrg::store_lock()
on the TABLE object of the other thread.

Changed ha_myisammrg::lock_count() and ha_myisammrg::store_lock()
to accept non-attached children. ha_myisammrg::lock_count() returns
the number of MyISAM tables in the MERGE table so that the memory
allocation done by get_lock_data() is done correctly, even if the
children become attached before ha_myisammrg::store_lock() is
called. ha_myisammrg::store_lock() will not return any lock if the
children are not attached.

This is however a change in the handler interface. lock_count()
can now return a higher number than store_lock() stores locks.
This is more safe than the reverse implementation would be.
get_lock_data() in the SQL layer is adjusted accordingly. It sets
MYSQL_LOCK::lock_count based on the number of locks returned by
the handler::store_lock() calls, not based on the numbers returned
by the handler::lock_count() calls. The latter are only used for
allocation of memory now.

No test case. The test suite cannot reliably run FLUSH between
lock_count() and store_lock() of another thread. The bug report
contains a program that can repeat the problem with some
probability.


include/myisammrg.h:
  Bug#30273 - merge tables: Can't lock file (errno: 155)
  Added mutex to struct st_myrg_info (MYRG_INFO).
sql/handler.h:
  Bug#30273 - merge tables: Can't lock file (errno: 155)
  Extended comments for handler::lock_count() and
  handler::store_lock().
sql/lock.cc:
  Bug#30273 - merge tables: Can't lock file (errno: 155)
  Changed get_lock_data() so that the final lock_count is taken
  from the number of locks returned from handler::store_lock()
  instead of from handler::lock_count().
sql/sql_base.cc:
  Fixed a purecov comment. (unrelated to the rest of the changeset)
storage/myisammrg/ha_myisammrg.cc:
  Bug#30273 - merge tables: Can't lock file (errno: 155)
  Changed ha_myisammrg::lock_count() and ha_myisammrg::store_lock()
  to accept non-attached children.
  Protected ha_myisammrg::store_lock() by MYRG_INFO::mutex.
storage/myisammrg/myrg_close.c:
  Bug#30273 - merge tables: Can't lock file (errno: 155)
  Added MYRG_INFO::mutex destruction to myrg_parent_close().
storage/myisammrg/myrg_open.c:
  Bug#30273 - merge tables: Can't lock file (errno: 155)
  Added MYRG_INFO::mutex initialization to myrg_parent_open().
  Protected myrg_attach_children() and myrg_detach_children()
  by MYRG_INFO::mutex.
  Fixed a purecov comment. (unrelated to the rest of the changeset)
2007-12-11 15:32:10 +01:00

68 lines
2.4 KiB
C

/* Copyright (C) 2000-2001 MySQL AB
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; version 2 of the License.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */
/* close a isam-database */
#include "myrg_def.h"
int myrg_close(MYRG_INFO *info)
{
int error=0,new_error;
MYRG_TABLE *file;
DBUG_ENTER("myrg_close");
/*
Assume that info->children_attached means that this is called from
direct use of MERGE, not from a MySQL server. In this case the
children must be closed and info->rec_per_key_part is part of the
'info' multi_alloc.
If info->children_attached is false, this is called from a MySQL
server. Children are closed independently but info->rec_per_key_part
must be freed.
Just in case of a server panic (myrg_panic()) info->children_attached
might be true. We would close the children though they should be
closed independently and info->rec_per_key_part is not freed.
This should be acceptable for a panic.
In case of a MySQL server and no children, children_attached is
always true. In this case no rec_per_key_part has been allocated.
So it is correct to use the branch where an empty list of tables is
(not) closed.
*/
if (info->children_attached)
{
for (file= info->open_tables; file != info->end_table; file++)
{
/* purecov: begin inspected */
if ((new_error= mi_close(file->table)))
error= new_error;
else
file->table= NULL;
/* purecov: end */
}
}
else
my_free((uchar*) info->rec_per_key_part, MYF(MY_ALLOW_ZERO_PTR));
delete_queue(&info->by_key);
pthread_mutex_lock(&THR_LOCK_open);
myrg_open_list=list_delete(myrg_open_list,&info->open_list);
pthread_mutex_unlock(&THR_LOCK_open);
VOID(pthread_mutex_destroy(&info->mutex));
my_free((uchar*) info,MYF(0));
if (error)
{
DBUG_RETURN(my_errno=error);
}
DBUG_RETURN(0);
}